Which stats command function provides a count of how many unique values exist for a given field in the result set?
This function of the stats command allows you to return the sample standard deviation of a field.
According to Splunk best practices, which placement of the wildcard results in the most efficient search?